Research map / Detection and segmentation / CNN features
Amplifying Discrepancies: Exploiting Macro and Micro Inconsistencies for Image Manipulation Localization
Shenghao Chen, Yibo Zhao, Tianyi Wang, Chunjie Ma, Weili Guan, Ming Li, Zan Gao
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| add ICM with single clustering step (ablation) | Segformer-b2 · CASIAv1/COVER/COCOGlide average | Permute-F1: 0.6176 -> 0.6350 | Improved |
| use full iterative clustering in ICM (ablation) | Segformer-b2 · CASIAv1/COVER/COCOGlide average | Permute-F1: 0.6350 -> 0.6680 | Improved |
| single-stage difference convolution instead of progressive DPM (ablation) | Segformer-b2 · CASIAv1/COVER/COCOGlide average | Permute-F1: 0.6176 -> 0.6004 | Hurt |
| use progressive DPM (ablation) | Segformer-b2 · CASIAv1/COVER/COCOGlide average | Permute-F1: 0.6004 -> 0.6386 | Improved |
| combine ICM and DPM (ablation) | Segformer-b2 · CASIAv1/COVER/COCOGlide average | Permute-F1: 0.6680 -> 0.6851 | Improved |
| combine ICM and DPM vs ICM alone (ablation) | Segformer-b2 · COCOGlide | Permute-F1: 0.6320 -> 0.6259 | Mixed |
| use FRD-Net vs Mesorch (fixed threshold) | FRD-Net · Average of 5 datasets | Permute-F1: 0.600 -> 0.688 | Improved |
| use FRD-Net vs Mesorch (optimal threshold) | FRD-Net · COVER | Permute-F1: 0.729 -> 0.717 | Hurt |
| use FRD-Net under Gaussian blur and noise vs Mesorch | FRD-Net · CASIAv1 perturbations | Permute-F1 | Hurt |
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.
- UnionFormer — UnionFormer: Unified-Learning Transformer with Multi-View Representation for Image Manipulation Detection and Localization (CVPR 2024) · better on 2 of 2 shared results
Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.
Closest papers by meaning
- Can We Model the Artifacts Explicitly? Disentangle Artifacts via Pairwise Edit Relations for Image Manipulation Localization (NeurIPS 2026 · similarity 0.87)
- Training-Free Image Manipulation Localization Using Diffusion Models (AAAI 2025 · similarity 0.87)
- Towards Modern Image Manipulation Localization: A Large-Scale Dataset and Novel Methods (CVPR 2024 · similarity 0.86)
- ForensicHub: A Unified Benchmark & Codebase for All-Domain Fake Image Detection and Localization (NeurIPS 2025 · similarity 0.86)
- Less Supervision, Better Generalization: Weakly Supervised Fake Region Localization in Diffusion-Edited Images (NeurIPS 2026 · similarity 0.86)
- Learning Discriminative Noise Guidance for Image Forgery Detection and Localization (AAAI 2024 · similarity 0.86)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.
